Probabilistic Flood Forecasting for Small Catchments using the G2G - PowerPoint PPT Presentation

Probabilistic Flood Forecasting for Small Catchments using the G2G Model Steve Cole, Alice Robson, Phil Howard, Vicky Bell and Bob Moore Centre for Ecology & Hydrology, Wallingford

Hydrological modelling Lumped and Distributed hydrological modelling Lumped Model Distributed Model (G2G) Gauging Gauging station • One model for each gauging station • One model for large regions (UK) • Many parameters calibrated to • Small set of regional parameters, observed flow location strong support from digital datasets • Flow estimates for one location only • Flow estimates in each grid (1km 2 ) • Uses catchment average rainfall • Uses gridded rainfall estimates

Need for distributed models of flood response Storm total Impact of spatial extent and location 435000 of storm on flood response? 430000 Distributed (G2G) Model Flow m 3 s -1 425000 420000 415000 355000 360000 365000 370000 375000 ― Catchment-wide storm 435000 Lumped Model ― Lower catchment storm 430000 Flow m 3 s -1 1 43 ― Upper catchment storm 425000 420000 415000 355000 360000 365000 370000 375000 435000 Hyetograph Darwen at Blue mm hr -1 mm 430000 Br. (135km 2 ) 425000 Moore et al. 420000 (2006), IAHS Pub. 305 415000 Time (days) 355000 360000 365000 370000 375000

Science Questions Motivation • Distributed hydrological models offer a natural approach to area-wide flood forecasting that includes small catchments BUT: • What rainfall estimates and forecasts should be • What rainfall estimates and forecasts should be used? • How to formulate area-wide distributed models for operational use in flood forecasting? • How do these area-wide models perform at small gauged and ungauged locations?

River Kent Case Study Gridded rainfall estimators: examples River Kent at • Using Hameldon Sedgwick (212km 2 ) Hill radar in North- West England • Two relatively • Two relatively steep upland catchments • Strong topographic control on flow response Darwen at Blue Cole and Moore Br. (135km 2 ) (2009), AWR

River Kent Case Study: Gridded rainfall estimators Gridded rainfall estimators: examples • River Kent catchment, orographic event, 3 Feb 2004 Gauge-adjusted 2km Nimrod Raingauge-only Raingauge River gauge 15 min totals • River Darwen catchment, convective event, 14 June 2002 Gauge-adjusted 1km raw radar Raingauge-only 15 min totals

River Kent Case Study: G2G model Grid-to-Grid distributed model (G2G) Evaporation Precipitation Surface flow-routing River Runoff- producing Saturation-excess soil column surface runoff River flow Return Drainage flow Subsurface flow-routing • Uses digital spatial datasets (e.g. terrain) • Responds to spatial variation of rainfall input • Grid-to-Grid routing using Kinematic Wave scheme

River Kent Case Study: G2G model G2G routing: use of terrain data 1. Flow directions : Elevation m apply automated 945 method to 50m DTM to infer 1km flow-paths 0 2. Catchment boundary delineation : inferred delineation : inferred from flow-path directions Sprint Mill Bowston Mint Bridge 3. Land/river designation: drainage area + river Victoria Bridge length threshold ! ! ! ! 4. Select forecast Sedgwick Sedgwick Sedgwick Sedgwick locations: gauged or ungauged

Gridded rainfall estimators G2G model assessment of rainfall estimators Novel multiquadric Improved rainfall Validated by surface fitting method estimates hydrological modelling (Cole & Moore, 2008) Gauge-adjusted radar 2km radar data Raingauge-only 15 min totals Distributed model (G2G) Observed Rating curve maximum

River Kent Case Study G2G model assessment at ‘ungauged’ sites Elevation m (70 km 2 ) (35 km 2 ) 945 Flow m 3 s -1 0 Sprint Mill Bowston Mint Bridge (66 km 2 ) (185 km 2 ) -1 Victoria Bridge Victoria Bridge Flow m 3 s - ! ! ! ! Sedgwick Sedgwick Sedgwick Sedgwick • G2G model calibrated at Sedgwick only (212 km 2 ) Flow m 3 s -1 Observed flow • 15-min raingauge data Model flow used Model baseflow Upper limit of • Comparable results at rating equation ungauged sites Time (days)

Flood forecasting using high-resolution NWP Hydrological flood forecasts using NWP Collaboration with the Joint Centre for Mesoscale Meteorology, EA and CEH using the Carlisle 2005 Floods (uses the PDM model). Q: Can new 1 or 4 km NWP rainfalls provide reliable flood forecasts? A: Yes, for the Carlisle floods (orographically enhanced frontal rain) 12 km (2005) 1 km (2011) Raingauge Caldew (246km 2 ) 4 km (2008) Flood Warning Level Flow m 3 s -1 Time (hours) Roberts, Cole et al., 12 km 4 km 1 km 2009, Met. Apps

Flood forecasting using high-resolution NWP Boscastle 2004 case study 4 km 12 km NIMROD radar Forecasts from 03 UTC 20 km radius Courtesy Nigel Roberts, from Boscastle JCMM (Met Office)

Flood forecasting using high-resolution NWP Boscastle 2004 case study • 1 or 4km NWP major improvement over 12km product • Still uncertainty in NWP rainfall intensities and location 1 km 12 km NIMROD radar Forecasts from 03 UTC 20 km radius Peak accumulations Courtesy Nigel Roberts, from Boscastle JCMM (Met Office) up to 50mm

Flood forecasting using high-resolution NWP Ensemble Flood Forecasting using G2G • Simple psuedo-ensemble method developed to capture NWP uncertainties. Genuine ensembles will be available in 2012(?) 1km NWP G2G Model 1km Comparison with river pseudo-ensemble river flow ensemble flow observations Acknowledgements: Collaboration with JCMM (Met Office)

Flood forecasting using high-resolution NWP Risk Map of flood exceedance using G2G ensembles and Q(T) flow return period grids Probability of exceeding a given flow threshold , for a given forecast horizon This example employs: This example employs: • NWP 1km rainfall pseudo- ensemble • 10 year return period flow thresholds • 24 hour forecast horizon 920 km 2 Potential to identify flood risk hotspots Acknowledgements: Collaboration with JCMM (Met Office)

Midlands Case Study – 20 July 2007 Ensemble 1 Ensemble 2 STEPS 6-hour spatial rainfall forecast 0900 to 1500 20 July 2007 20 ensembles Avon & Tame (Midlands) catchments STEPS Zero rainfall HyradK 6 hrs (padding out) raingauge Ensemble 3 Ensemble Average 130 km 2 130 km 2 Radar Composite HyradK raingauge T+0 T+6 Ensemble hydrographs observed flow modelled flow using raingauge rainfall 20 flow ensembles using STEPS Ensemble average rainfall is zero rainfall used beyond 6hr STEPS less than raingauge rainfall but higher than radar

Midlands Case Study – 20 July 2007 G2G ensembles using STEPS forecasts 20 6-hour STEPS ensemble rainfall forecasts in G2G 20 July 2007 93 km 2 185 km 2 Observed Modelled 20 STEPS ensemble 93 km 2 91 km 2 members members 20 July 2007 Traditional 74 km 2 130 km 2 ensemble outputs at gauged locations

Midlands Case Study – 20 July 2007 Probability of exceedance flood maps T+3 hours T+6 hours 10 year return period flood threshold 6-hr STEPS forecasts then zero rainfall 20 STEPS Members 09:00 20 July 2007 origin Avon & Tame (Midlands) catchments 75-100% 75-100% Key indicates probabilities of (number of 50-75% members) exceeding the 10-year flood . 25-50% 10-25% T+12 hours T+18 hours 2-10% During early part of storm, highest exceedance probabilities are on the very small rivers. As time progresses the main exceedance hotspots are on the larger rivers and can be tracked moving downstream and meeting at confluences.

National application of G2G National application of G2G • Several EA/Defra R&D projects recommended nationwide operational trial of G2G for flood forecasting – 2004-06: Extreme Event Recognition Phase 2 (FD2208) – 2005-07: Rainfall-runoff and other modelling for ungauged/low- benefit locations (SC030227) – 2007-10: Hydrological modelling using convective scale rainfall – 2007-10: Hydrological modelling using convective scale rainfall modelling (SC060087) • Pitt Review of the Summer 2007 floods – Environment Agency/Met Office Flood Forecasting Centre (FFC) for England & Wales, opened April 2009 – Scottish Flood Forecasting Service (SFFS) between SEPA/Met Office opened 2010 – G2G now undergoing operational trials in FFC and SFFS

National application of G2G G2G runoff production: use of soil property associations Runoff production key element – needs to reflect heterogeneous soil properties Use of Soil Survey data (HOST, Seismic, other…) to obtain 1km grids of: • water content at field capacity • residual soil water content • porosity • saturation hydraulic conductivity • horizon depth Association Issues: table links Scale 29 HOST soil Effective values classes to soil Bell et al. (2009), JoH Lateral properties properties Moore et al. (2006), IAHS Pub. 305

National application of G2G G2G national application • G2G runs nationally within NFFS/FEWS using a 15 min time- step and models river flow and soil moisture on a 1km grid • Ongoing operational trial and assessment Raingauge-adjusted River flow Soil moisture radar

69 km 2 Examples of catchments with 53 km 2 generally good G2G performance January & February 2008 93 km 2 93 km 2 Observed Modelled 357 km 2 357 km 352 km 2 Demonstrates modelling of different flow regimes and catchment sizes with the G2G Model 191 km 2 9,962 km 2